catenary-mcp 1.1.0

# Configuration

Catenary loads configuration from multiple sources, in order of priority (last one wins):

1.  **Defaults**: `idle_timeout = 300`, `smart_wait = true`.
2.  **User Config**: `~/.config/catenary/config.toml`.
3.  **Project Config**: `.catenary.toml` in the current directory or any parent directory (searches upwards).
4.  **Explicit File**: Specified via `--config <path>`.
5.  **Environment Variables**: Prefixed with `CATENARY_` (e.g., `CATENARY_IDLE_TIMEOUT=600`).
6.  **CLI Arguments**: `--lsp` and `--idle-timeout`.

## Basic Structure

```toml
# Global settings
idle_timeout = 300  # Seconds before closing idle documents (0 to disable)
smart_wait = true    # Wait for server initialization before first request

# Language servers
[server.<language-id>]
command = "server-binary"
args = ["arg1", "arg2"]
```

## JSON Schema

A JSON schema is available in the repository at `catenary-config.schema.json`. You can use this to get autocompletion and validation in editors like VS Code.

To use it in VS Code, add this to your `settings.json`:

```json
"yaml.schemas": {
  "https://raw.githubusercontent.com/MarkWells-Dev/Catenary/main/catenary-config.schema.json": [".catenary.toml", "catenary.toml"]
}
```
*(Note: Requires the YAML extension which also handles TOML schemas in some versions, or use a dedicated TOML extension that supports `$schema` comments).*

## Example Config

```toml
idle_timeout = 300

[server.rust]
command = "rust-analyzer"

[server.rust.initialization_options]
check.command = "clippy"

[server.python]
command = "pyright-langserver"
args = ["--stdio"]

[server.typescript]
command = "typescript-language-server"
args = ["--stdio"]

[server.javascript]
command = "typescript-language-server"
args = ["--stdio"]

[server.go]
command = "gopls"

[server.php]
command = "php-language-server"
```

## Initialization Options

Each server can receive custom `initialization_options` that are passed to the
LSP server during the `initialize` request. These are server-specific settings
that configure the server's behavior.

```toml
[server.rust]
command = "rust-analyzer"

[server.rust.initialization_options]
check.command = "clippy"
cargo.features = "all"
```

Refer to your language server's documentation for available options.

## Language IDs

The `[server.<language-id>]` key must match the LSP language identifier. Catenary detects these based on file extension and some common filenames:

| File / Extension | Language ID |
|------------------|-------------|
| `.rs` | `rust` |
| `.py` | `python` |
| `.ts` | `typescript` |
| `.tsx` | `typescriptreact` |
| `.js` | `javascript` |
| `.jsx` | `javascriptreact` |
| `.go` | `go` |
| `.c` | `c` |
| `.cpp`, `.cc`, `.cxx`, `.h`, `.hpp` | `cpp` |
| `.cs` | `csharp` |
| `.java` | `java` |
| `.kt`, `.kts` | `kotlin` |
| `.swift` | `swift` |
| `.rb` | `ruby` |
| `.php` | `php` |
| `.sh`, `.bash`, `.zsh` | `shellscript` |
| `Dockerfile` | `dockerfile` |
| `Makefile` | `makefile` |
| `CMakeLists.txt`, `.cmake` | `cmake` |
| `.json` | `json` |
| `.yaml`, `.yml` | `yaml` |
| `.toml`, `Cargo.toml`, `Cargo.lock` | `toml` |
| `.md` | `markdown` |
| `.html` | `html` |
| `.css` | `css` |
| `.scss` | `scss` |
| `.lua` | `lua` |
| `.sql` | `sql` |
| `.zig` | `zig` |
| `.mojo` | `mojo` |
| `.dart` | `dart` |
| `.m`, `.mm` | `objective-c` |
| `.nix` | `nix` |
| `.proto` | `proto` |
| `.graphql`, `.gql` | `graphql` |
| `.r`, `.R` | `r` |
| `.jl` | `julia` |
| `.scala`, `.sc` | `scala` |
| `.hs` | `haskell` |
| `.ex`, `.exs` | `elixir` |
| `.erl`, `.hrl` | `erlang` |

## Global Options

| Option | Default | Description |
|--------|---------|-------------|
| `idle_timeout` | `300` | Seconds before auto-closing idle documents. Set to `0` to disable. |
| `smart_wait` | `true` | Wait for LSP servers to be fully ready before executing requests. |

## Shell Execution (`tools.run`)

The `run` tool allows AI agents to execute shell commands. It is **disabled by
default** — add a `[tools.run]` section to enable it.

```toml
[tools.run]
allowed = ["git", "make"]  # Always allowed commands

# Language-specific commands (activated when matching files exist in workspace)
[tools.run.python]
allowed = ["python", "pytest", "uv"]

[tools.run.rust]
allowed = ["cargo"]
```

**Recommended base allowlist for constrained mode:**

```toml
[tools.run]
allowed = ["chmod", "cp", "diff", "git", "gh", "ln", "make", "mkdir", "mv", "rm", "rmdir", "touch", "wc"]
```

This covers file management, version control, and basic utilities without
opening up arbitrary shell access. Avoid adding commands that can write
arbitrary content to stdout (`cat`, `echo`, `printf`, `bash`) — these bypass
Catenary's file tools and their path validation.

**Key behaviors:**

- Commands not on the allowlist are **rejected** with an error showing the
  current allowlist.
- Set `allowed = ["*"]` for **unrestricted** execution (use with caution).
- Language-specific commands activate automatically when matching files are
  detected in the workspace (e.g., `.py` files activate the `python` group).
- The tool description updates dynamically to reflect the current allowlist.
- Commands are executed **directly** (not via shell) to prevent injection.
- Output is capped at 100KB per stream. Default timeout is 120 seconds.

## CLI Override

You can also specify servers via CLI:

```bash
catenary --lsp "rust:rust-analyzer" --lsp "python:pyright-langserver --stdio"
```